OpenAI: GPT-3.5 Turbo Instruct

Text input Text output
Author's Description

This model is a variant of GPT-3.5 Turbo tuned for instructional prompts and omitting chat-related optimizations. Training data: up to Sep 2021.

Key Specifications
Cost
$$$$
Context
4K
Parameters
175B
Released
Sep 27, 2023
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Logprobs Response Format Temperature Max Tokens Structured Outputs Presence Penalty Top Logprobs Stop Logit Bias Top P Seed Frequency Penalty
Features

This model supports the following features:

Structured Outputs Response Format
Performance Summary

OpenAI's GPT-3.5 Turbo Instruct, released on September 27, 2023, is a variant of GPT-3.5 Turbo optimized for instructional prompts. This model performs among the fastest models, consistently ranking in the 82nd percentile across five benchmarks. It offers moderate pricing, placing in the 38th percentile. Notably, the model demonstrates exceptional reliability with a 99% success rate, indicating minimal technical failures and consistent evaluable responses. In terms of performance across benchmarks, GPT-3.5 Turbo Instruct shows a mixed profile. It achieves strong accuracy in General Knowledge (93.5%) and Ethics (97.0%), though these scores place it in the 32nd and 29th percentiles respectively, suggesting room for improvement compared to top-tier models. Its Coding performance is fair at 71.0% accuracy (28th percentile). A significant weakness is observed in Instruction Following, where it scores only 38.0% accuracy (30th percentile). Furthermore, its ability to acknowledge uncertainty and avoid hallucinations is limited, with a 60.0% accuracy in the Hallucinations benchmark (12th percentile). Key strengths include its high speed and outstanding reliability, making it a dependable choice for applications where consistent and rapid responses are critical. Its primary weakness lies in its instruction following and hallucination mitigation capabilities.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $1.5
Completion $2

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
OpenAI
OpenAI | openai/gpt-3.5-turbo-instruct 4K $1.5 / 1M tokens $2 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by openai